Timeline for On modifying our scope: A proposal
Current License: CC BY-SA 3.0
20 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 19, 2012 at 12:33 | history | edited | KavehMod | CC BY-SA 3.0 |
added 261 characters in body; added 20 characters in body; added 17 characters in body; added 15 characters in body
|
| Aug 19, 2012 at 12:22 | history | edited | KavehMod | CC BY-SA 3.0 |
added 286 characters in body; added 24 characters in body; added 68 characters in body
|
| Aug 12, 2012 at 17:54 | comment | added | vzn | "working at bell labs 2 decades I learned that questions from practitioners are often vague/illinformed but if you talk with them you very often get clearer questions which are often quite interesting". reminds me of an old semifamous steve jobs quote, one of the worlds greatest product designers (in application of theory): "its not the customers job to know what they want".. also maybe paraphrased as "the customer doesnt know what they want" | |
| Aug 11, 2012 at 18:37 | comment | added | vzn | @peter -- aggressive yet inappropriate "guilty until proven innocent" attitude is spot on. had two application-of-theory questions closed/downvoted aggressively recently. & kaveh the tag warrior didnt add it. didnt notice that tag until now.. ouch, its perfect for a lot of my questions. would edit them to chg the tag, but am sure its not gonna help the votes. live & learn, 2020 hindsight. unfortunately kaveh seems to act like tags on questions are highly objective and not so subjective, see him declare strongly against tags I thought were highly appropos. guess some tags are "ghetto" | |
| Aug 11, 2012 at 13:33 | comment | added | Jeffε | @Raphael: Maybe I should say "Otherwise, I assume it's not." Not that I'll insist that it's not. | |
| Aug 11, 2012 at 10:55 | comment | added | Raphael | @JɛffE: That's not a very good heuristic (for maintaining the site), as it does not transfer to others (i.e. mods) and probably does not work outside of your area of expertise/teaching. | |
| Aug 10, 2012 at 15:04 | comment | added | Jeffε | Here's my own judgement about what constitutes a "homework question": If I have asked that precise question on an undergraduate algorithms exam, it's a homework-level question. If I've given the question in an undergrad algorithms homework, and most of the students got it, it's a homework-level question. If it strikes me as a perfect question to ask the undergrad algorithms class I'm currently teaching, then it's probably a homework-level question. Otherwise, it's not. | |
| Aug 9, 2012 at 19:35 | comment | added | Jeffε | @PeterShor: I don't agree that "How do you modify SMAWK to do X?" and "How do you do X?" are the same question! (Also, SMAWK is (sadly) not covered in most algorithms classes, so it's not clear even the former question would be closed.) | |
| Aug 9, 2012 at 17:55 | comment | added | Kaveh Mod | I think asking for motivation in unclear cases and waiting for a day or two is a good middle ground. How much we are following this in practice is another issue. We can also add an item to our FAQ for practitioners (though I seriously doubt anyone reads the FAQ these days). We also have this comment. | |
| Aug 9, 2012 at 17:44 | comment | added | Kaveh Mod | @Peter, I understand your concerns and do find them valid. My main concern here is to have a workable solution, everyone has good intentions but we should also consider the general consequences of these decisions. Are we going to leave these questions open indefinitely even if the OP doesn't engage? How many undergrad questions will pass as unclear cases if we do so? Would that make the scope meaningless? What will be the effect of these on the participation of researchers? | |
| Aug 9, 2012 at 17:32 | comment | added | Peter Shor | Look at it from the OP's point of view. His question, on a stackexchange site he doesn't understand that well, has just been rejected by a bunch of really smart people who are generally university professors. Is he going to argue with us? | |
| Aug 9, 2012 at 17:03 | comment | added | Kaveh Mod | @Peter, I think we have already agreed that application of theory questions are on-topic (IIRC, as long as we don't get too many). Closing is not permanent, the general practice is that when a question is not fit it is closed, when the question is edited to become fit it is reopened. I would not call this "guilty until proven innocent", we tell the OP that the question is not fit, if the OP engages and improves the question fine, if after sometime there is no reply we close the question until the question is improved to become fit. And we also tell these to the OP using comments/FAQ. | |
| Aug 9, 2012 at 16:54 | comment | added | Kaveh Mod | @Peter, is that an application of theory question? It seems more like a programming/implementation question to me, have a look at the other question the same user posted around the same time. cstheory is not a coq help forum as far as I know. (An analogy for how I see the question: how can I fix the following C++ code, I am getting compile error X on g++). If you think the question should be reopened please start a new meta discussion. IIRC I closed them after 1 day or so and no reply from the OP to the comments. | |
| Aug 9, 2012 at 15:41 | comment | added | Peter Shor | I think the problem is that we have the "guilty until proven innocent" rather than "innocent until proven guilty" mindset. If a question is clearly homework, or clearly from somebody who wants to understand something about theoretical computer science they should learn from reading Wikipedia or an undergrad textbook, we should close it. Otherwise, we should leave it open. Working at Bell Labs for nearly two decades, I learned that questions from practitioners are typically vague and ill-formed, but if you talk with them, you very often get clearer questions which are often quite interesting. | |
| Aug 9, 2012 at 15:32 | comment | added | Peter Shor | This question for example, clearly wasn't homework, and so should not have been closed so precipitously, even if it wasn't very well-formed. | |
| Aug 9, 2012 at 15:18 | comment | added | Kaveh Mod | @Peter, I think the issue is that we cannot know if it is from a practitioner unless the OP tells us it is so. If you notice a case please add the application-of-theory tag and flag it for moderator attention and we can reopen the question. | |
| Aug 9, 2012 at 15:16 | comment | added | Peter Shor | Not if people are currently closing questions from practitioners for being at too low a level. | |
| Aug 9, 2012 at 15:13 | comment | added | Kaveh Mod | @Peter, we had a discussion about that also IIRC, and I think the result was adopting the application-of-theory tag. Doesn't this address your concern? | |
| Aug 9, 2012 at 15:09 | comment | added | Peter Shor | I think a "question at the level of undergraduate homework assignment/textbook exercises" should be off-topic if it comes from somebody who encountered it as a textbook exercise. Let's take a variant of the SMAWK algorithm. If the homework exercise is "modify the SMAWK algorithm to solve this problem", it might be easy. If somebody who has never heard of the SMAWK algorithm is trying to develop an algorithm for the same problem, it's not at all easy, and in fact this problem might stump CS researchers who last heard of the SMAWK algorithm in a class 10 years ago. And it's the same question. | |
| Aug 9, 2012 at 14:49 | history | answered | KavehMod | CC BY-SA 3.0 |